|
Student
Abstracts at NREL:
Acoustic Array for Wind Turbine Noise Analysis. CHRISTOPHER
BONILHA and IAN TSE (Univeristy of Colorado, Boulder, CO,
80309) PATRICK MORIARTY (National Renewable Energy Laboratory,
Golden, CO, 89401)
Locating and
characterizing sources of noise from wind turbines can greatly
aid in the design and production of quieter, more publicly accepted
machines for renewable power generation. An acoustic array is a
device comprised of an arrangement of microphones that when coupled
with an algorithm, can locate sources of noise. In 2006, The National
Renewable Energy Laboratory (NREL) partnered with the University
of Colorado at Boulder to construct a prototype acoustic array
as a proof of concept. Issues arose in both the original hardware
and software components which needed troubleshooting and correction
before the capabilities of the array could be determined. Tests
showed that erroneous signals being outputted by the array were
caused by the original data acquisition
(DAQ) hardware’s inability to handle the high volume of data samples. A robust,
differential-referenced, simultaneous-sampling DAQ was purchased to replace the
old DAQ, resolving the data acquisition issues. The low-quality microphones had
inconsistent frequency responses that contributed to the erroneous results. It
was also determined that the signal-to-noise ratio could be significantly improved
with better microphone arrangements and with the doubling of the number of microphones
on the array. The beamforming algorithm that computes the sound pressure levels
emanating from a given plane of interest was originally written incorrectly and
very inefficiently. A new program was written to perform the beamforming algorithm
on the recorded audio signals and produce plots to facilitate easier analysis.
Simulations were performed to analyze how array parameters contribute to the
performance of the array. After hardware upgrades and the software revisions,
the array was subjected to a series of simulations and tests to determine its
capabilities. The array was unable to detect a monopole sound source roughly
4 meters away. Further tests should be done on an array with more microphones
of better quality and also with a source that is both louder and positioned at
probable turbine locations.
Alternative Fuels Data Center: Fleet Reports, Databases and Website Redesign. BRETT HOAG (University of Colorado, Boulder, CO, 80309)
JOHANNA LEVENE (National Renewable Energy Laboratory, Golden,
CO, 89401)
The Alternative
Fuels Data Center (AFDC) is online collection of data, including
more than 3,000 documents and several interactive tools. The AFDC
collaborates
with the U.S. Department of Energy’s (DOE) Clean Cities Program as well as the
Energy Policy Act of 1992 (EPAct) fleet programs. Federal fleet location reports
for EPAct were processed using excel, geocoding software and cgi scripts that
evaluated the received data to the data located in the AFDC (AFDC) Alternative
Fuel Station Locator Database. A total of 15,267 unique fleet locations were
processed resulting in an addition of 73,841 vehicles to the AFDC Database. The
updating process of the Related Links Database used several different debugging
procedures and techniques, as well as work with Oracle database maintenance software.
The Clean Cities Success Stories Database update process required extensive work
contacting organizations, writing and editing summaries about organizations that
are currently implementing alternative fuels within their fleet. During the updating
process 131 individual organizations were contacted, resulting in a new Success
Stories Database that had 15 new organization summaries as a base. Debugging
procedures and techniques were also applied to several online tools available
for fleets as well as the public. The Make/Model Application, Flex-Fuel Fleet
Vehicle Cost Calculator and the AFDC Laws and Incentives page were debugged for
potential problems that new users may experience. The research was conducted
to aid in the development of the new AFDC website that has a planned launch date
of September 30, 2007.
Analysis of Renewable Energy Deployment in Colorado by 2030. RUSSELL
MUREN (University of California at Berkeley, Berkeley, CA, 94701) CHUCK
KUTSCHER (National Renewable Energy Laboratory, Golden, CO, 89401)
Currently most utilities
in the state of Colorado are subject to the 20% renewable portfolio standard
(RPS) passed by voters in 2004 and expanded by the state legislature in 2007.
However, because of bonuses and exemptions written into the law, the true required
renewable energy penetration is only 12.3%. This makes this law less then adequate
for addressing climate change. This study aims to assess the real renewable energy
and carbon impacts of the current RPS and investigates the benefits of increasing
the RPS to true 20% and 30% values. To this end a user input-driven predictive
Excel model was developed to find the proper technology spread, electrical outputs,
and carbon reduction for each RPS. It was found that while all the RPS variants
are technically feasible based on available renewable resources, only the 30%
RPS meets the carbon reductions that are thought necessary to avoid the worst
impacts of climate change. Based on the results of this report the current RPS
does not offer an effective avenue to reduce fossil fuel and carbon reduction.
Furthermore, if the goal of the current Colorado legislature and administration
is carbon reduction, a 30% RPS is the most acceptable avenue.
Applied Material and Energy Evaluation for Biomass Gasification. ISAAC SACHS-QUINTANA (New Mexico Institute of Mining
and Technology, Socorro, NM, 87801) CALVIN FEIK (National Renewable Energy
Laboratory, Golden, CO, 89401)
As gasoline prices continue
to climb, and the price of corn increases, the thermochemical conversion of lignocellulosic
biomass into fuels becomes more economical. Economic modeling is the primary
tool for assessing the feasibility of biomass conversion processes. Material
and energy balances on experimental data are needed to validate the theoretical
economic models. Material and energy balances were performed on a pilot scale
biomass gasification plant. The plant’s gasifier, thermal cracker, and tar reformer
were considered. Usable data was extracted from the pilot plant’s data acquisition
system. Differentiation and integration were performed to achieve a basis of
calculation. Input and output flows were compared, and material and energy closures
were executed. The average material closure for the gasifier, thermal cracker,
and tar reformer were 74.34% ± 28.77%, 97.71% ± 3.129%, 100.1% ± 15.05%, respectively.
The average energy closures for the thermal cracker and the tar reformer are
107.9% and 101.3% respectively. At the moment, an energy closure for the gasifier
cannot be determined because of insufficient data. The experimental data for
the tar reformer and thermal cracker can readily be used for validating economic
models. The gasifier data cannot. Additional tests in the pilot plant are required
to obtain more accurate material and energy streams.
Assembly of a Time-Correalted Single Photon Counting Experimental Setup. SEAN SWEETNAM (Carleton College, Northfield, MN, 55057)
RANDY ELLINGSON (National Renewable Energy Laboratory, Golden, CO, 89401)
Recent developments in nanotechnology
have created materials capable of improving the efficiency of solar cells, provided
the basic photophysics involved are well understood. To properly characterize
and understand the charge carrier processes which occur in nanomaterials, it
is necessary to use very fast light gathering techniques, with time resolutions
as small as tens of picoseconds. Time-Correlated Single Photon Counting (TCSPC)
is such a technique, capable of providing sufficiently fast time resolution to
resolve important process in materials by utilizing fast response mechanisms
of detectors and electronic components, and by using efficient trigger timing.
TCSPC also extends the range of observable photoluminescence with its long wavelength
and low power detection capabilities. The goal of the work discussed in this
paper is to develop a TCSPC system unique in its time resolution and range of
detection. In this paper the principles and components of TCSPC are described,
and the preparation of a TCSPC experimental setup is discussed, in particular
noting the systematic errors encountered and their solutions. The system was
run with a Tsunami Ti:Sapphire laser operating at 864-868 nm with a Si Avalanche
Photodiode (APD) detector, yielding a time resolution of less than 800 ps. Emission
lifetime measurements of 3.6 nm diameter PbS quantum dots with this setup yield
a lifetime greater than 1300 ns; this value varied for different emission wavelengths
965-1030 nm. Because the time resolution is more than three orders of magnitude
shorter than the lifetime of PbS quantum dots, it is concluded that the system
is sufficiently fast for typical carrier lifetime characterization. Further development
of the system will be necessary to improve the time resolution and infrared capabilities
of the system; in particular the inclusion of an InGaAs APD in the system will
decrease the time resolution, and increase the detection range to as far as 1600
nm. A flexible setup permitting fast switches between the InGaAs and Si detectors
will increase the usefulness of the setup by increasing the full range of sensitivity
to 400-1600 nm. Further experimentation will be necessary to determine the cause
of the emission lifetime
variation associated with emission wavelength.
Blends of thiophene-based dendrimers with titania nanoparticles for use in
organic photovoltaic devices. TALIA
GERSHON (Massachusetts Institute of Technology, Cambridge,
MA, 2139) DAVID GINLEY (National Renewable Energy Laboratory,
Golden, CO, 89401)
Dendrimers
(small branched organic molecules) offer an exciting alternative
to typical semiconducting polymers in organic photovoltaic applications,
as they are monodisperse and have virtually independently tunable
optical, electronic, and architectural properties. Similarly,
metal oxide nanoparticles offer a tunable, high-mobility alternative
to the established organic electron acceptors. Since nanoparticles
are many times larger than typical organic acceptors they may
also help address issues in domain size limitations that appear
in fullerene-based bulk heterojunctions. To explore these advantages
a unique blend of the thiophene-based dendrimer, 4G1-3S, and
anatase titania nanoparticles has been studied for use in organic
photovoltaic devices. The charge transfer between these two materials
was monitored as a function of solvent, drying conditions, and
device architecture. Changes in film domain size and morphology
were observed via SEM and AFM images. Devices were made under
a variety of conditions via solution processing in air. Although
no photocurrent was consistently extracted, several key conclusions
were drawn regarding how these materials interact, which point
toward the realization
of a successful device.
Cd2SnO4 and ZnMgO: Ternary Transparent Conducting Oxides for Thin Film CdTe
and CuInGaSe2 Solar Cells. HANNAH RAY (Wesleyan University,
Middletown, CT, 6459) XIAONAN LI (National Renewable
Energy Laboratory, Golden, CO, 89401)
The compounds
cadmium stannate (Cd2SnO4)
and zinc magnesium oxide
alloy (Zn1-xMgxO) are ternary transparent conducting oxides (TCOs) that are
promising for use in thin film solar cells. When used in place of the commonly
used TCO tin oxide (SnO2), Cd2SnO4 enhances small-scale cadmium telluride (CdTe) solar cell performance.
To scale up the deposition area and rate of Cd2SnO4,
metal organic chemical vapor deposition is used. The commonly used precursors
for cadmium oxide and
SnO2 have
incompatible deposition temperatures; thus, a different tin oxide precursor with
a lower deposition temperature must be found to be used with dimethyl cadmium.
A literature search was performed, and tin (II) acetylacetonate was deemed best
for this application. Zinc oxide (ZnO) is commonly used in copper indium gallium
diselenide (CIGS) solar cells. Alloying ZnO with the insulator magnesium oxide
(MgO) allows for
band gap (Eg) engineering. However, adding MgO to a ZnO alloy detracts from
the film’s electrical conductivity. In this study, a gradient of ZnO:Al and MgO
was deposited using RF magnetron sputtering by two ZnO:Al and MgO targets, creating
the zinc magnesium oxide alloy Zn1-xMgxO. Magnesium (Mg) content of the films varied from 3-30%. The
optical and electronic properties of the film were measured every centimeter
along the gradient. It was determined that the electrical properties of the Zn1-xMgxO
alloys are only acceptable for use in a solar cell when the proportion of Mg
in the sample is less than
3.5% (xg
(Zn1-xMgxO) = 1.72x + Eg
(ZnO) with the amount of Mg in the alloy. At Mg content between 0-7%,
the band gap was higher than this fit line due to the Burstein-Moss effect. As
Mg content increased in this region, Eg due to the Burstein-Moss effect decreased until the effect disappeared
at x = 0.07. Increasing the deposition temperature improved the electronic conductivity
of the films. Increasing the temperature had no effect on the band gap when the
Mg content was greater than 7%, but when Mg content was less than 7%, Eg increased with temperature.
Characterization and Performance of the Zonal Exposure to Broadband RAdiation
(ZEBRA) Shadowband. SARAH BRADEN (Northwestern
University, Evanston, IL, 60201) DARYL MYERS (National
Renewable Energy Laboratory, Golden, CO, 89401)
Cost-effective
measurement of solar radiation resources is a worldwide problem.
The Zonal Exposure to Broadband RAdiation (ZEBRA) shadowband was
developed by Michael J. Brooks at the University of KwaZulu-Natal
(UKZN) in Durban, South Africa, to reduce the cost of providing widespread
data for solar resource assessment and climate change. The shadowband
consists of regularly spaced perforations, allowing alternate measurement
of diffuse and global radiation throughout the day. A pyranometer
equipped with the ZEBRA shadowband is used to independently measure
diffuse and global irradiance. Clear sky direct irradiance may be
calculated without relying on a moving shadowband or other instruments.
With the aid of an absolute cavity radiometer, or other reference
pyrheliometer, the ZEBRA can provide shade-unshade calibration for
pyranometers. This project characterizes and investigates the accuracy
of the ZEBRA shadowband for data acquisition and pyranometer calibration.
Data from National Renewable Energy Laboratory (NREL) Solar Radiation
Research Laboratory (SRRL) and UKZN was used to develop an algorithm
to reconstruct global and diffuse clear sky profiles, test the shade-unshade
pyranometer calibration method, derive and verify responsivity values,
investigate thermal offset correction, calculate a shadowband correction
factor, perform an uncertainty analysis, compare derived and reference
data, study site dependence (using experimental results from southern
and northern-hemisphere trials), and investigate performance under
partly cloudy and overcast conditions. The combined statistical (2-sigma)
and bias calculated uncertainty is about +/-
17 W/m2 for diffuse irradiances and +/- 50 W/m2 for
global and direct irradiances. Comparisons of ZEBRA data with reference data
have average empirical
uncertainty of approximately +/- 30 W/m2 or better for direct estimates, +/- 20 W/m2 for
global, and +/- 15
W/m2 for
diffuse. Our results demonstrate that the ZEBRA has the potential for use in
mainstream radiation resource assessment as an alternative to multiple expensive
instruments. The ZEBRA concept also has the potential to work for other types
of radiometers. Further work may include the development of a software package
for processing
ZEBRA data as discussed here.
Characterization of a Mobile Oscillatory Fatigue Operator for Wind Turbine
Blade Testing. PEARL DONOHOO (Franklin W Olin College of Engineering,
Needham, MA, 2492) JASON COTRELL (National Renewable Energy Laboratory,
Golden, CO, 89401)
Laboratory testing of wind
turbine blades is required to meet wind turbine design standards, reduce machine
cost, and reduce the technical and financial risk of deploying mass-produced
wind turbine models. Fatigue testing at the National Wind Technology Center (NWTC)
is currently conducted using Universal Resonance Excitation (UREX) technology.
In a UREX test, the blade is mounted to a rigid stand and hydraulic exciters
mounted to the blade are used to excite the blade to its resonant frequency.
A drawback to UREX technology is that mounting hydraulic systems to the blade
is difficult and requires a relatively long set-up period. The author has analyzed
an alternative testing technology called the Mobile Oscillatory Fatigue Operator
(MOFO). The MOFO uses an oscillating blade test-stand rather than a rigid stand,
avoiding the need to place hydraulic systems on the blade. The MOFO will be demonstrated
by converting an existing test-stand at the NWTC to an oscillating stand that
can test blades up to 25m in length. To obtain the loads necessary to design
the MOFO, the system motion is modeled using rigid body and lumped mass dynamics
models. Preliminary modeling indicates the existing stand can be converted to
a MOFO relatively easily. However, the blade dynamic models suggest that blade
bending moment distributions are significantly different for UREX and MOFO testing.
More sophisticated models are required to assess the implication of this difference
on the accuracy of the test.
Characterization of GaN, In025Ga075N and In050Ga050N for Photoelectrochemical
Water Splitting. SALLY PUSEDE (University of Colorado at Denver
and Health Sciences Center, Denver, CO, 80217) TODD DEUTSCH
(National Renewable Energy Laboratory, Golden, CO, 89401)
GaN, In0.25Ga0.75N
and In0.50Ga0.50N
semiconductors were characterized as possible candidates for photoelectrochemical
water splitting. The band gap energies of the materials were measured
and found to decrease with increasing indium content. The flatband
potential (Vfb) of GaN was determined and the material’s band edges
verified to span the potentials of the hydrogen and oxygen evolution
reactions;
however,
the Vfb positions of In0.25Ga0.75N and In0.50Ga0.50N could not be experimentally established. Two-electrode current
density vs. potential measurements of both InGaN materials indicated anodic current
flow at zero applied potential considerably below theoretical maxima, suggestive
of photocorrosion rather than spontaneous water splitting. Where GaN was observed
to be stable, In0.25Ga0.75N and In0.50Ga0.50N were found to be extremely susceptible to photocorrosion.
Characterization of Silicon Carbide for Water-Splitting and Corrosion Protection. BORIS CHERNOMORDIK (University
of Louisville, Louisville, KY, 40292) DR. JOHN TURNER (National Renewable Energy Laboratory, Golden,
CO, 89401)
Harnessing hydrogen as an energy carrier
by means of renewable power, such as solar, is key to achieving a clean and dependable
energy cycle. Direct photoelectrolysis of water with photoelectrochemical (PEC)
cells is the most efficient method to collect hydrogen from water. In this study,
commercial samples of the 4H polytype of silicon carbide (SiC), with n- and p-type
doping, were photoelectrochemically characterized with a focus on durability
for possible use as a protective coating for chemically sensitive high-efficiency
PEC materials. The indirect band gap was found to agree with the literature value
of 3.23eV and the bands were found to straddle the water electrolysis redox potentials,
with the flat band potential vs. pH relationship exhibiting Nernstian behavior.
Corrosion experiments showed that n-type SiC is susceptible to oxidative etching,
while p-type SiC did not show signs of corrosion. Though the high band gap renders
4H-SiC inefficient for photoelectrolysis because it absorbs only a tiny portion
of the solar spectrum, corrosion experiments suggest that p-type SiC may be durable
enough to act as a protective layer in PEC cells. More experimentation in this
regard is necessary, though. In addition, more research is needed into methods
for incorporating SiC as a protective coating for a high conversion efficiency
cell.
Combinatorial Study of Znx Sn1-x Oy: a Transparent Conducting Oxide. DIANA SILVA (University of Colorado, Boulder, CO,
80301) MAIKEL VAN HEST (National Renewable Energy Laboratory, Golden,
CO, 89401)
Zinc-tin-oxide (ZTO)
libraries were
co-sputtered from ZnO and SnO2 by RF magnetron sputtering onto Eagle 2000 glass substrates
at temperatures ranging from room temperature
to 550°C and analyzed with combinatorial analytical techniques. Samples were
deposited in pure argon and in a mixture of argon and hydrogen with 4% and 2%
hydrogen. ZTO films were found to be amorphous at a tin content ranging from
30 to 70% when deposited in pure argon gas at 550°C. ZTO films were also amorphous
when deposited in a mixture of argon/hydrogen with
4% and 2% hydrogen at 550°C. ZTO films deposited in pure argon had a maximum
conductivity of 134 S ⁄cm at a 70% tin content and an optical transparency
with transmission 85%. While, ZTO films deposited in a mixture of argon and hydrogen
2% had an optical transparency with transmission 85% across the visible spectrum
and a maximum conductivity of only 92 S ⁄cm observed at 75 at % tin content.
Thin films of zinc-tin-oxide deposited in a mixture of argon and hydrogen with
4% and 2% hydrogen, did not improve conductivity and optical properties as expected.
Key words: zinc tin oxide, transparent conducting oxides, amorphous
Conceptual Design of a New Large Scale Wind Turbine Drive Train Testing Facility. SCOTT LAMBERT (University of Colorado, Boulder, CO,
80303) JASON COTRELL (National Renewable Energy Laboratory, Golden,
CO, 89401)
Laboratory testing of wind
turbine drive trains is an important way to validate designs, test reliability,
debug systems, and verify computer and analytical models. The large physical
size and high torque requirements of modern wind turbines present engineers
with unique manufacturing and testing challenges. The 2.5 MW drive train testing
facility at the National Renewable Energy Laboratories (NREL) in Colorado is
one of a few facilities capable of testing multi-megawatt wind turbine drive
trains. The
rapid growth of wind turbine size has outpaced the
facility’s capacity to test very large wind turbine systems. The goal of the
research described in this report is to identify possible configurations, assess
the technical challenges, and investigate the costs for a new, larger, 12MW drivetrain
test facility. The identification of potential dynamometer configurations was
conducted by examining large test facilities in use overseas, and through consultation
with industry. Several conceptual designs were modeled using computer aided design
software and a preliminary engineering analysis for each concept was conducted.
Readily-available components that could be used in this project and suppliers
capable of assisting with engineering and manufacturing of limited production
components were identified. The concepts where then judged against each other
on the basis of cost, component availability, and relative ease of implementation.
This research shows that configurations using large custom built gearboxes and
motors are expensive and require long lead times due to technical obstacles in
engineering and manufacturing. Concepts using large custom components show higher
overall system costs than those using multiples of smaller, more readily-available
components that distribute the high torque loads over several load paths. While
initial estimates show that these distributed-load systems have considerable
potential for cost savings, further investigation into these concepts is required
to assess the risks involved. Furthermore, the results of this study indicate
that concepts such as single motor and gearbox combinations suitable for use
in smaller scale test bench systems do not scale well up to 12MW, and do not
offer the same setup and configuration flexibility that is possible with distributed
load concepts.
Cross-Sectional Polishing for Solar Cell Device Characterization. NATHAN
FAST (California State University Northridge, Northridge, CA, 91330)
BHUSHAN SOPORI (National Renewable Energy Laboratory, Golden, CO, 89401)
Fabrication of high efficiency
solar cells requires a method to study the effects of new processes on the
interactions and physical structure within a cell. An improved method for cross-sectional
polishing of large areas of solar cells is described. This method produces
a highly planar surface compared to the conventional way of preparing a sample
by cleaving. The traditional method of cleaving or fracturing a sample works
well for single crystalline wafers, however, creates undesirable surface morphologies
for polycrystalline wafers or multi-layered devices such as finished solar
cells. The discontinuities caused by the fracture mechanics of non-homogeneous
samples create patterns that make it difficult to characterize the true nature
of the materials under study. Solar cells cross-sectioned by this improved
technique can be characterized by very high resolution electron-beam and optical
imaging to measure the alloyed regions of front and back contacts, thickness
of backsurface field, and other important physical properties of solar cells.
In this method, a standard polycrystalline Si solar cell sample is prepared
in a specially designed polishing chuck and secured with wax. The sample is
sequentially mechanically polished by progressively decreasing the polishing
grit size with chemical mechanical polishing (CMP) used as the last step. Optical
microscope and scanning electron microscope (SEM) micrographs showing the planar
results of this progressive polishing method are presented showing the improvement
over the cleaving technique. This large area cross-sectioning permits statistically
significant evaluation of many areas of the cell. The planarity of the sample
edge makes it possible to perform a variety of atomic force microscope (AFM),
conductive atomic force microscope (CAFM), scanning Kelvin probe microscope
(SKPM) and other scanning analyses over large areas in addition to more localized
investigations
such as with SEM.
Cytochemical Investigation of Lignin Redistribution During Thermochemical Pretreatment. BRITNEY PENNINGTON (Florida Institute of Technology, Melbourne, FL, 32901)
TODD VINZANT (National Renewable Energy Laboratory, Golden, CO, 89401)
Due to the increasing demand
for oil, the United States has developed starch ethanol programs, but corn cannot
support both the food and fuel industries. Cellulosic ethanol is a promising
alternative to starch-based ethanol but is more difficult to generate cost-effectively
because biomass is inherently resistant to degradation. Lignin, the polyphenolic
compound in plant cell walls, contributes to this recalcitrance by inhibiting
hydrolytic cellulases and presents an obstacle to producing bioethanol. Dilute
acid pretreatment of biomass removes only a fraction of the lignin content, and
yet at high temperatures, sufficient enzymatic digestion can still occur. To
address this paradox, this study utilized microscopy and cytochemical stains
to determine temperature’s role in lignin redistribution during dilute acid pretreatment.
All of the cytochemical stains used to detect lignin had evenly distributed staining
patterns at 80ºC but became concentrated towards the cell edges as temperatures
approached 160ºC. Temperature’s effect on the biomass surface was also investigated
using scanning electron microscopy.
Starting at 140ºC, half-sphere droplets appeared on the tissue surfaces and their
morphologies seem to coalesce into larger spheres at higher temperatures. Round
droplets were also observed using the light microscope. It has been hypothesized
that the melted lignin is pushed out of the cell wall, possibly by increased
hydrogen bonding between adjacent cellulose microfibrils, and forms spheres due
to hydrophobic forces. Understanding lignin redistribution and its resulting
implications on cell and tissue structure will help biologists explain the effects
of pretreatment on biomass.
Degradation of Organic Light Emitting Diodes. ERIC ELLENOFF
(Stanford University, Stanford, CA, 94305) JOSEPH BERRY (National Renewable
Energy Laboratory, Golden, CO, 89401)
Organic light emitting
diodes (OLEDs) currently degrade much faster than standard inorganic light
emitting
diodes.
In order to improve OLEDs’ lifetimes, it is necessary to measure their degradation
in a controlled environment to determine what mechanism is responsible. A measurement
setup was constructed to characterize
the decay of an OLED’s IV curve, emission intensity, and emission spectrum. A
baseline for device performance and degradation was established by measuring
their performance in a glovebox where they were surrounded by Argon with water
and oxygen kept below .1ppm. A second set of devices was tested outside of the
glovebox in a prototype for a portable encapsulation device designed and built
at NREL known as a “puck”. By comparing the degradation of these devices to the
set which was kept in the glovebox, we were able to determine how effective the
puck was at protecting OLEDs. Over the course of a 165-hour test, the intensity
of the OLED tested in the glovebox fell by 62% which indicates a half life of
115.5 hours. The relative emission of light at each wavelength remained constant
throughout this test. The OLED which was tested in a puck degraded much more
rapidly: it lost half of its intensity after every 19 hours of use, and degraded
even when not in use. Furthermore, as the OLED became dimmer, its emission spectrum
shifted, resulting in significantly less light being emitted between 540nm and
660nm. For these reasons, the puck proved to be an inadequate for encapsulating
devices. A second-generation test would feature the ability to leak controlled
concentrations of reactants such as oxygen and water vapor to the device.
Increasing the Accuracy of Global Irradiance Calculations: An Analysis of Responsivities
and Correction Methods. LIZA BOYLE
(University of the Pacific, Stockton, CA, 95211) STEPHEN WILCOX (National
Renewable Energy Laboratory, Golden, CO, 89401)
In an effort to make solar
radiation data more accurate for solar energy system and climate change research
there have been many advances in solar radiometer calibration leading to the
creation of several different pyranometer responsivities and correction methods.
Here
we study the accuracy of four responsivities - "responsivity
(45º)", "responsivity function", "responsivity (45º) corrected",
and "responsivity function corrected" - and three correction methods
- "Reda" (which relates directly to the two corrected responsivities), "Dutton",
and "Full." Data was gathered from the National Renewable Energy Laboratory
and Atmospheric Radiation Measurement program, Southern Great Plains sites, over
a two and half year period. The average difference, or deltas, between a reference
irradiance, determined from independent direct and diffuse irradiances, and the
irradiance calculated from pyranometer data using these different methods was
examined. The averages and standard deviations of these deltas indicate the accuracy
and precision of the pyranometer data. Analysis of data showed that the "responsivity
function" reduced the zenith angle dependence apparent in the "responsivity
(45º)", decreasing the overall standard deviation of the deltas from 15.15
W/m^2 to 11.74 W/m^2. Average deltas decreased from 4.40 W/m^2 to 1.46 W/m^2
by using the "responsivity function". Analysis also showed that "responsivity
(45º) corrected" slightly decreased the average delta of the "responsivity
(45º)" data from 4.40 W/m^2 to 1.56 W/m^2, while keeping the scatter relatively
constant, 15.15 W/m^2 to 13.45 W/m^2 respectively. The "responsivity function
corrected" slightly increased the average delta of the "responsivity
function" data from 1.46 W/m^2 to 1.51 W/m^2, while keeping the scatter
relatively constant, 11.74 W/m^2 to 12.31 W/m^2 respectively. When applied to "responsivity
(45º)" data, the "Dutton" and the "Full" methods reduced
the average delta from 4.40 W/m^2 to 0.23 W/m^2 and 0.97 W/m^2 respectively,
but increased scatter from 15.15 W/m^2 to 16.26 W/m^2 and 16.13 W/m^2 respectively.
These results indicate that that "responsivity function" and "responsivity
function corrected" have the greatest accuracy and least uncertainty. Further
studies are needed to understand why the "Dutton" and "Full" methods
increase scatter, understand all of the trends revealed in the data, and compare
other responsivities and correction methods with those analyzed in this study.
Inkjet Printing of Nickel and Silver Metal Solar Cell Contacts. ROBERT PASQUARELLI (Rochester Institute of Technology, Rochester, NY, 14623)
CALVIN CURTIS (National Renewable Energy Laboratory, Golden, CO, 89401)
With about 125,000 terawatts
of solar power striking the earth at any given moment, solar energy may be the
only renewable
energy resource with enough capacity to meet a major portion of our future energy
needs. Thin-film technologies and solution deposition processes seek to reduce
manufacturing costs in order to compete with conventionally coal-based electricity.
Inkjet printing, as a derivative of the direct-write process, offers the potential
for low-cost, materials-efficient deposition of the metals for photovoltaic contacts.
Advances in contact metallizations are important because they can be employed
on existing silicon technology and in future-generation devices. We report on
the atmospheric, non-contact deposition of nickel (Ni) and silver (Ag) metal
front contacts from metal-precursor organic inks on a Dimatix inkjet printer
at 180-220°C. Near-bulk conductivity Ag contacts were successfully printed up
to 4.5 µm thick and less than 125 µm wide on the silicon nitride antireflective
coating of silicon solar cells. Thin, high-resolution Ni adhesion-layer lines
were printed on glass and zinc oxide at 55 nm thick and
80 µm wide with a conductivity two orders magnitude less than bulk. Additionally,
the ability to print multi-layered metallizations (Ag on Ni) on transparent conducting
oxides was demonstrated and is promising for contacts in copper-indium-diselenide
(CIS) solar cells. Future work will focus on further improving resolution, printing
full contacts on devices, and investigating copper inks as a low-cost replacement
for Ag contacts.
In-Situ Stress Measurement for MOCVD Growth of High Efficiency Lattice-Mismatched
Solar Cells. ALEJANDRO LEVANDER (Pennsylvania
State University, State College, PA, 16802) JOHN GEISZ (National
Renewable Energy Laboratory, Golden, CO, 89401)
Dislocations, formed in
order to relieve stress, act as sites for nonradiative electron/hole pair recombination,
which reduces the efficiency of photovoltaics. Stress forms as a result of
mechanical and thermal mechanisms during the metal-organic chemical vapor deposition
growth process. Mechanical stress is the result of depositing lattice-mismatched
(LMM) layers on top of one another and thermal stress results from a thermal
gradient within the sample and depositing layers with different thermal expansion
coefficients on top of one another. To reduce the number of dislocations in
the active layer when using LMM materials, compositionally step-graded layers
and a buffer layer are placed between the two LMM materials. In order to achieve
a better understanding of the effect of stress on the active layer, the buffer
composition, and therefore lattice constant, was varied to change the stress
on the active layer. The in-situ stress and ex-situ strain were characterized
using a multi-beam optical stress sensor and x-ray diffractometry respectively.
The quality of the photovoltaic devices was measured using a solar simulator
and quantum efficiency instrument. Samples with near zero stress or small amounts
of compressive stress in the active layer had the highest open-circuit voltages
and efficiencies. Tensile stress in the active layer significantly degraded
performance. The biaxial modulus was calculated from a stress v. strain curve,
but several sources of error exist. The band gap of the active layer increased
with increasing stress, despite the composition remaining constant. Future
work will concentrate on the effect of dopant type on stress development and
dislocation
formation in the graded layer.
Isolation and Analysis of Above Cloud Point Precipitates in Palm and Poultry
Fat Derived Biodiesels. REBECCA CALLAHAN (Hendrix College, Conway, AR, 72032) TERESA ALLEMAN (National Renewable Energy Laboratory, Golden,
CO, 89401)
As the search for alternatives
to petroleum based fuels continues many promising options have arisen. Vegetable
and animal
fat derived biodiesels have become one viable alternative to petroleum based
diesel fuel. Biodiesel can be directly substituted for petrodiesel while exhibiting
less hazardous and superior environmental properties, such as its high flash
point and biodegradability. Biodiesel typically has high cloud and pour points
than conventional diesel. Additionally, a small fraction of biodiesels also show
a precipitate formed above the cloud point. This precipitate may accumulate in
storage tanks or on fuel filters inhibiting
biodiesel’s distribution. The identification of this precipitate may assist in
finding a solution to its operability problems or an identification method of
the fuels it impairs, thus enhancing the capability of biodiesel as an alternative
fuel. Multiple samples of biodiesel were visually examined for the presence of
precipitate at room temperature and two samples of different feed stocks were
chosen for precipitate analysis, one palm and one poultry fat. Three steps were
implemented to identify their precipitates. First, the mass of precipitate was
increased by chilling biodiesel above its cloud point. The precipitate was then
isolated from biodiesel through filtration and solvent washing. Lastly it was
identified using the following analytical techniques: Fourier Transform Infrared
Spectroscopy (FT-IR), Gas Chromatography Flame Ionization Detector (GC-FID) pyrolysis
Molecular Beam Mass Spectrometry (py-MBMS), and proton, carbon 13 and Distortionless
Enhancement by Polarization Transfer Nuclear Magnetic Resonance (H NMR and 13C
NMR and DEPT NMR.) Using these methods of analysis it was determined that the
two samples differed in physical appearance, yet both contained a large quantity
of monoglycerides. Approximately ten percent of the samples examined in this
work showed precipitation upon cooling, additional examination of biodiesels
needs to be conducted to verify that this precipitation is universal to all feed
stocks. Additionally, after the definitive identification of the precipitates,
their impact on fuel filters must be quantified from the distributor to the vehicle.
Mechanization of Cadmium Sulfide Chemical Bath Deposition. DOMINIC
WEBER (Colorado School
of Mines, Golden, CO, 80401) STEVEN ROBBINS (National Renewable Energy
Laboratory, Golden, CO, 89401)
The preferred method to deposit the
Cadmium Sulfide (CdS) layer for thin-film solar cell research is Chemical Bath
Deposition (CBD). In order to improve the reproducibility, and thus help the
understanding of the process, a machine was created to mimic the CBD process.
The design of the system involved placing the substrate over a flow well, only
depositing CdS on one side of the substrate. Within the machine was a pump, designed
to make the chemicals flow at a designated speed across the substrate. Also included
was a temperature bath, designed to heat the chemicals to the desired deposition
conditions. The machine was constructed, and everything in the system was tested
for functionality. Since the machine was designed to mimic the current CBD method,
the processed films were compared with CdS films made by the old process. These
tests involved: thickness, thickness uniformity, and optical properties. The
first run of the system revealed some major uniformity issues: the film had areas
of no deposition (bubbles) on the edges and the thickness varied quite a bit
on the other areas of the film. The first run also revealed another major problem:
waste minimization. The run required
2.5 L (for 16 in2 of substrate) compared to 500 mL (for 13.5 in2)
in the old CBD method. The second run was devoted to waste minimization. A beaker
was placed in the temperature bath to displace volume, reducing the amount of
required chemicals
to 1.5 L (for 36 in2). The second run showed that a smaller amount of liquid still
produced a good quality film. Further designs for the CBD machine will have to
concentrate on making the flow, and therefore the film thickness uniform. Future
designs will also have to keep in mind the waste management issue: a great design
would be one that uses 500 mL of solution or less.
Optimization of Spray-Coated Organic Photovoltaics. RENEE
GREEN (University of Pittsburgh, Pittsburgh, PA, 15261) GARRY RUMBLES
(National Renewable Energy Laboratory, Golden, CO, 89401)
Organic photovoltaic
devices, traditionally spin-coated, were made via an airbrush spray-deposition
technique from a 1:1
poly(3-hexylthiophene) (P3HT) and [6,6]-phenyl-C61 butyric
acid methyl ester (PCBM) blend in dilute (1mg/ml each) solution. Devices were
tested for dependence on solvent boiling point and annealing temperature. Working
devices resulted from solutions prepared in chloroform, toluene, chlorobenzene,
and para-xylene, with an initial maximum power conversion efficiency
of over 2% (average 1.787%) from a p-xylene solution. Chlorobenzene was selected for use in further
studies due to its small statistical spread of device efficiencies, high-quality
smooth films, and comparable (1.15%) efficiency values. Annealing the device
active layer at 120°C resulted in the highest power conversion efficiency among
annealing temperatures ranging from 90°C to 200°C. Spray-coating is ideal for
its ability to deposit highly dilute solutions, create multilayer organic devices,
and expand the range of available analysis techniques by permitting the creation
of high-quality thick films.
Plug-in Hybrid Electric Vehicle and Hybrid Electric Vehicle Emissions under
FTP and US06 Cycles at High, Ambient, and Low Temperatures. MATTHEW
SEIDMAN (California State Polytechnic University, Pomona, CA, 91768)
TONY MARKEL (National Renewable Energy Laboratory, Golden, CO, 89401)
The concept of a Plug-in
Hybrid Electric Vehicle (PHEV) is to displace consumption of gasoline by using
electricity from the vehicles large battery pack, to power the vehicle as much
as possible with minimal engine operation. This paper assesses the PHEV emissions
and operation. Currently, testing of vehicle emissions is done using the federal
standard FTP4
cycle on a dynamometer at ambient (75°F) temperatures. Research was also completed
using the US06 cycle. Furthermore,
research was completed at high (95°F) and low (20°F) temperatures. Initial dynamometer
testing was performed on a stock Toyota Prius under the standard FTP4 cycle,
and the more demanding US06 cycle. Each cycle was run at 95°F, 75°F, and 20°F.
The testing was repeated with the same Prius retrofitted with an EnergyCS Plug-in
Hybrid Electric system. The results of the testing confirm that the stock Prius
meets Super-Ultra Low Emission Vehicle requirements under current testing procedures,
while the PHEV Prius under current testing procedures were greater than Super-Ultra
Low Emission Vehicle requirements, but still met Ultra Low Emission Vehicle requirements.
Research points to the catalyst temperature being a critical factor in meeting
emission requirements. Initial engine emissions pass through with minimal conversion
until the catalyst is heated to typical operating temperatures of 300-400°C.
PHEVs also have trouble maintaining the minimum catalyst temperature throughout
the entire test because the engine is turned off when the battery can support
the load. It has been observed in both HEVs and PHEVs that the catalyst is intermittently
unable to reduce nitrogen oxide emissions, which causes further emission releases.
Research needs to be done to combat the initial emission spikes caused by a cold
catalyst. Research also needs to be done to improve the reduction of nitrogen
oxides by the catalyst system.
Polyhydroxyl fullerene with silicotungstic acid hydrate to create a composite
and integrate with Nafion ® polymer to create a proton exchange membrane. JOSHUA LAU (Colorado School of Mines, Golden, CO,
80401) JOHN A. TURNER (National Renewable Energy Laboratory, Golden,
CO, 89401)
The U.S. Department of Energy’s
(DOE)
goal is to have a membrane for the PEM
that will operate at 120-150 °C
and a relative humidity (RH) of 25% or less by the year 2010. Proton exchange
membranes (PEMs) for fuel cell applications have current limitations on the materials
and capabilities of the membrane as is. There has been research to modify the
perfluorosulfonic acid (PFSA) polymer based membranes, most notably, Nafion ®, by doping PFSA membranes with different compounds, such as
heteropoly acids (HPAs), to increase the conductivity of the membranes, and with
other particles to improve the longevity of the membranes at high temperatures
and low relative humidity. In this research, a combination of polyhydroxyl fullerene
(PHF), a fullerene derivative, and silicotungstic acid hydrate (HSiW), were combined
in two separate solutions and mole ratios of one-to-one and one-to-four, respectively,
to create a composite. Infrared spectroscopy (IR) was conducted on the composite
at various stages of the process. The results showed that the resulting compound
differed from the starting material. An in-situ method of mixing of the two compounds
on a stir plate in mole equivalent ratios of one-to-one, one-to-two, one-to-three,
and one-to-four in the DuPont DE2820 polymer matrix was done in order to make
membranes. Three controls were also cast at the same time to get base conductivities
for comparison. The conductivities of the in-situ mixed membranes of the controls
were found to have conductivity values of 132 ± 15 mS·cm -1,
23.4 ± 0.2 mS·cm -1, and 156 ± 15 mS·cm -1 for PHF; 129 ± 18 mS·cm -1, 21.5 ± 3.0 mS·cm -1, and 150 ± 21 mS·cm -1 for HSiW; and 130 ± 8 mS·cm -1, 24.2 ± 4.4 mS·cm -1, and 162 ± 2 mS·cm -1 for DuPont DE2820. The mole ratio conductivities had average
values 146 ± 20 mS·cm -1, 26.6 ± 22 mS·cm -1, and 194 ± 20 mS·cm -1. All of the values were
at 60°C 100%RH, 80ºC 50%RH, and 80°C 100%RH, respectively. The values are promising
compared to Nafion ® which
is well known and reported. It is also interesting that added separately, the
materials actually had worse conductivity than that of the pure DuPont DE2820,
but when the materials
were combined, the conductivity was greater in all cases.
Preliminary form and footprint studies for the integrative design of NREL Research
Support Facility. MICHAEL SMITH (Portland State
University, Portland, OR, 97201) PAUL TORCELLINI (National Renewable
Energy Laboratory, Golden, CO, 89401)
Commercial buildings currently
account for 18% of annual U.S. energy consumption. The U.S. Department of Energy
(DOE) has established the goal of achieving market viable commercial zero-energy
buildings by 2025. A net zero-energy building (ZEB) is a building with significantly
reduced energy demand through efficiency improvements such that the remaining
energy needs can be supplied with onsite renewable technologies. Building design
problems are inherently multivariate and multi-objective, encompassing a large
number of parameters to consider. Balances must be researched and found to maximize
an overall net gain in performance. A whole-building or integrative design approach
takes into account the interactions among the subsystems of a building and requires
collaboration through interdisciplinary design teams which includes key players
throughout the building process. On March 16th, 2007, the DOE approved $73 million
to design and construct a new administrative building termed the Research Support
Facility (RSF). DOE and NREL view this as an opportunity to create a national
showcase how aggressive energy efficiency goals can be achieved. The studies
covered in this report are intended to provide recommendations for key players
involved in the process of designing the RSF. The studies are focused on early-phase
preliminary design decisions. The emphasis is on examining the effects of building
form and footprint on the energy performance of a building. This assessment is
simulation-based, systems integration analysis using NREL’s Opt-E-Plus software
based on DOE’s EnergyPlus building simulation program to model the energy performance
of different options for the RSF. The results are analyzed to develop recommendations
for optimal building shapes that take into account trade offs between energy
use for heating, cooling and lighting. The recommended floor plate is roughly
rectangular with an aspect ratio of 2.5, floor-to-floor height of 12.5 ft, and
the longer walls of the building facing north and south.
Meeting the RSF’s goal of energy use intensity of 25 kBtu/ft2•yr should be possible
as long as typical energy efficient measures are applied, plug loads do not exceed
0.5 W/ft2, and installed lighting loads are around 0.7 W/ft2.
Silicon Nitride for Semiconductor Photoelectrochemical Water Splitting. JOSEPH RYERSON (University
of Colorado, Boulder, CO, 80309) JOHN TURNER (National Renewable Energy Laboratory, Golden,
CO, 89401)
Silicon nitride was analyzed to determine
if it is an effective material to split water in direct photoelectrolysis application.
Band gaps determined via photocurrent spectroscopy were found to exhibit indirect
transitions between 1.54-1.75 eV. The higher end nitrogen content samples (14-16%
nitrogen) fell within the band gap range effective in splitting water (1.7-2.2
eV). Flat band potentials were not negative enough to drive the hydrogen evolution
reaction for these n-type films. Silicon nitride is therefore ineffective as
a direct photoelectrolysis device, because its band edges are not aligned with
respect to the redox potentials, a criteria necessary to generate hydrogen from
water splitting. Twenty-four hour durability tests in 1M KOH revealed that this
material is highly stable, as determined from post-test optical microscopic analysis,
with low nitrogen contents (5%), and much less stable at higher nitrogen contents
(14-16%N).
Thermochemical Ethanol via Indirect Gasification of Lignocellulosic Biomass
with Methanol and Dimethyl Ether Intermediates. MICHELLE
HARRIS (Colorado School of Mines, Golden, CO, 80401) STEVEN PHILLIPS
(National Renewable Energy Laboratory, Golden, CO, 89401)
Thermochemical gasification
is a process where a carbonaceous feed is partially oxidized to a gas-phase fuel
(syngas)
consisting mostly of carbon monoxide (CO) and hydrogen (H2) gases. Syngas can
be converted to fuels such as ethanol, methanol, or dimethyl ether (DME) via
a catalytic process called synthesis. A 2007 report by S. Phillips, A. Aden,
J. Jechura, and D. Dayton at the National Renewable Energy Laboratory (NREL)
provided a detailed techno- economic evaluation of a 2,000 tonne wood to ethanol
process via gasification and synthesis. The present study evaluates a new scenario
that consists of five of these small plants producing methanol or DME from wood,
and a larger plant producing ethanol via the methanol or DME collected from the
smaller plants. Achieving improved economy of scale for the large ethanol production
plant is the reasoning behind this new study. The mass and energy calculations
needed for this study were done using the computer
simulation program ASPEN®, which models chemical and industrial processes, with
the original ASPEN simulation designed by Phillips et al. modified to meet the
new specifications. The cost of production (COP) of each of the products (methanol,
DME, or ethanol) was determined to assess the feasibility of the new scenario
versus the original process. The COP was calculated using a Discounted Cash Flow
Rate of Return (DCFROR) method programmed in Microsoft® EXCEL. The COP, in units
of $/MM Btu and $/gal ethanol equivalent, ranged from $8.73-$10.20/MM Btu and
$0.67-$0.78/gal for the methanol and DME scenarios. The methanol to ethanol process
used in this study produced ethanol at a COP of $1.26/gal, which is significantly
higher than the $1.01/gal from the Phillips et al. study. Future research will
continue to work on the scenarios involving DME and other process configurations
to optimize the industrial plants to achieve their greatest efficiency and to
decrease the cost of production.
Well-to-Wheel Analysis of Renewable Fuels in Hybrid and Plug-In Hybrid Vehicles. CHRISTINE RYAN (University of Colorado, Boulder, CO,
80310) CHRISTINE RYAN (National Renewable Energy Laboratory, Golden,
CO, 89401)
Well-to-Wheel (WTW) analyses
have been used for a number of years to understand the energy and environmental
impacts
of various types of fuel and vehicle systems.
As the issues of reducing America’s dependence on petroleum and decreasing transportation
sector emissions become increasingly important so do WTW analyses on the vehicle
and fuel systems of the future in order to understand appropriate pathways. Plug-In
Hybrid Electric Vehicle (PHEV) technology and renewable biomass fuels have the
potential to significantly reduce the petroleum consumption of the transportation
sector and in turn the United States as a whole. Using the Greenhouse gases,
Regulated Emissions, and Energy use in Transportation (GREET) model developed
by Argonne National Laboratory, WTW analyses were conducted for 12 vehicle/fuel
systems. Baseline spark-ignition (SI) and baseline compression-ignition (CI)
engines were compared to HEV and PHEV vehicles using various fuels; reformulated
gasoline (RFG), low-sulfur diesel (LSD), 20% biodiesel blend (BD20), and 85%
ethanol blend (E85). The result of using E85 and B20 shows an over all reduction
in the use of fossil fuels, in both HEV and PHEV models when compared to the
baseline RFG vehicle, and this reduction is larger when renewable fuels are used.
The largest reduction was seen with a HEV vehicle paired with E85 fuel. PHEV
vehicles also saw reductions in fossil fuel use, with a PHEV40 reducing the fossil
fuel use more when compared to an HEV using RFG fuel. PHEVs have greater petroleum
reductions than HEV vehicles with all fuel types used. Emissions from PHEV and
HEV models differ as fuels change. The greatest reduction in CO2 emissions
was seen with E85 fuel used in an HEV, with a reduction of 58.7% .
|
